Your Trusted Partner for Audit Analytics Fraud Workshop Finding the truth in the transactions Copyright 2011 ACL Services Ltd. Robin Clough, ACDA ACL Certified Trainer
Copyright 2011 ACL Services Ltd. 2 Your Trusted Partner for Audit Analytics Presenter Agenda ACL and fraud detection Creating a fraud detection program Features in ACL that can assist with fraud detection Sample analytics / Analytics in AX Questions
Copyright 2011 ACL Services Ltd. 3 ACL and fraud detection
Copyright 2011 ACL Services Ltd. 4 ACL and fraud detection Trawl large volumes of data Evidence could be hidden behind millions of transactions Don t rely on lottery approach, focus on 100% of the transactions Use ACL profiling commands to find trends/outliers Use ACL filtering to home in on unusual transactions
Copyright 2011 ACL Services Ltd. 5 ACL and fraud detection Span more than one dataset Bring several sets of data into the analysis from different systems Use ACL combining commands to cross compare data
Copyright 2011 ACL Services Ltd. 6 ACL and fraud detection Use a continuous approach Look at implementing AX Fraud occurs on an ongoing basis Run analytics on a regular basis Ensure results are reviewed Exception Ensure analytics are kept fresh and current
Copyright 2011 ACL Services Ltd. 7 Creating a fraud detection program
Copyright 2011 ACL Services Ltd. 8 Creating a fraud detection program Build a profile of potential frauds Identify high risk areas Look at historical fraud Inside/outside organisation What s happening elsewhere media, industry groups
Copyright 2011 ACL Services Ltd. 9 Creating a fraud detection program Develop analytics to look for possible indicators of fraud Use full breath of ACL features Convert to scripts Refine and repeat over time
Copyright 2011 ACL Services Ltd. 10 Creating a fraud detection program Improve controls by implementing continuous controls monitoring Don t wait too long to detect the fraud it could be too late Best caught in early stage before real damage is done
Copyright 2011 ACL Services Ltd. 11 Features in ACL that can assist with Fraud Detection
Copyright 2011 ACL Services Ltd. 12 Features in ACL that can assist with Fraud Detection STATISTICS Command Gives averages, value or date spans Lots of values stored as variables that could be used in filters and scripts
Copyright 2011 ACL Services Ltd. 13 Features in ACL that can assist with Fraud Detection DUPLICATES Command Easy to use, key is in selection of data and computed fields Common application is in analysis of duplicate payments Suppliers may be deliberately double invoicing to exploit control weaknesses Could also be collusion between AP and Suppliers
Copyright 2011 ACL Services Ltd. 14 Features in ACL that can assist with Fraud Detection CLASSIFY / SUMMARIZE Commands Great for spanning large datasets and seeing the data in a concise way Take vendor creation/modification Is the role of K_DUDAK really to create/modify vendors? Why blanks? Use in combination with AVERAGE1 variable generated by STATISTICS Set tolerance boundaries e.g. 20% below AVERAGE1 Sometimes good to view graphically also
Copyright 2011 ACL Services Ltd. 15 Features in ACL that can assist with Fraud Detection STRATIFY Command Good way to quickly see outliers in a large population of numbers Could be used when viewing expenses
Copyright 2011 ACL Services Ltd. 16 Features in ACL that can assist with Fraud Detection FIND Function Easy to locate key strings of data in large data sets Is not case sensitive and also looks within strings Famous Gym/Taxi scam
Copyright 2011 ACL Services Ltd. 17 Features in ACL that can assist with Fraud Detection BENFORD Command Probability theorem that makes predictions about distribution of digits in numerical data Can detect hidden trends occurring in the data Looks at first few digits of numbers and attempts to predict expected distribution Used frequently to detect spending just below authorisation levels Always worth running to see what it comes up with anything
Copyright 2011 ACL Services Ltd. 18 Features in ACL that can assist with Fraud Detection BENFORD Command
Copyright 2011 ACL Services Ltd. 19 Features in ACL that can assist with Fraud Detection FILTERS Allows you to home in on transactions that meet certain conditions Quick filter allows expressions to be built quickly with no syntax errors Use variables and functions to make your filters powerful Any guesses as to what this filter is doing?
Copyright 2011 ACL Services Ltd. 20 Features in ACL that can assist with Fraud Detection COMPUTED FIELDS / WORKSPACES Add your own fields to existing data fields Confirm accuracy of calculations Store filters as logical computed fields i.e. T or F For widely used computed fields think about using Workspaces
Copyright 2011 ACL Services Ltd. 21 Features in ACL that can assist with Fraud Detection SCRIPTS Why reinvent the wheel every time? Share/swap your scripts with your friends! Easier than you think to create recorder, log entries, create script from table history What s this script doing?
Copyright 2011 ACL Services Ltd. 22 Sample Analytics
Copyright 2011 ACL Services Ltd. 23 Sample Analytics Purchase to Pay Split purchases Inflated prices (collusion) Duplicate payments Appropriate authorisation Invoice amount doesn t match PO amount
Copyright 2011 ACL Services Ltd. 24 Sample Analytics Purchase to Pay Cross check between employees/contractors and suppliers Check key information is not fictitious e.g. address details, VAT number Segregation of duty analysis roles matrix Vendor creation and modification
Copyright 2011 ACL Services Ltd. 25 Sample Analytics Purchase cards Merchant profiling Personal use e.g. holiday travel instead of business travel! Weekend use, outside business hours Split purchases (Benford) Peer group comparison Restricted merchants
Copyright 2011 ACL Services Ltd. 26 Sample Analytics Travel and entertainment expenses Duplicate claims No receipts Inappropriate use Wrong rates Policy breach e.g. business travel v economy travel, meal allowance exceeded Peer group comparison Authorisation checks Benford
Copyright 2011 ACL Services Ltd. 27 Sample Analytics Journals Raised during non working days or hours Isolate manual journals Created by senior management Large amounts / Small amounts Round sum amounts Benford analysis
Copyright 2011 ACL Services Ltd. 28 Sample Analytics Others Timesheet/Swipe card analysis Point of Sale analysis Stock levels/wastage levels
Copyright 2011 ACL Services Ltd. 29 Analytics in AX
Copyright 2011 ACL Services Ltd. 30 Questions?